26 BULLETIN 1328, U. S. DEPARTMENT OF AGRICULTURE 



as when a swarm is on the wing. When the curve for this day's 

 record was plotted, an unusual peak, about 40 per cent higher than 

 the normal trend of the curve, was noticeable at this time on both 

 the ingoing and the outgoing curves. Although other peaks occur 

 at different times, usually only on a single curve at one time, none was 

 so marked as this particular one. If this peak was actually a re- 

 sponse to this disturbance in the neighboring colony, it is interest- 

 ing to speculate as to the method of conveyance of the information 

 to the bees within the experimental colony. This question suggests 

 another use for these instruments, namely, a study of the reaction 

 of bees to odor, sound, and other external stimuli. 



THE AVERAGE DURATION OF TRIPS 



A study of the factors which influence the length of time the bees 

 are absent from the hive on each foraging trip is a distinct problem 

 in itself. The amount of nectar available in the flowers, the position 

 and nature of the nectaries, and the distance which the bees must 

 travel to obtain this nectar are the chief factors which govern the 

 average duration of the voyages; therefore, an intimate knowledge 

 of this duration and its variations is of importance in a study of 

 nectar secretion, especially with reference to the time of day when 

 the maximum secretion occurs. At present it is possible to give 

 only a limited selection of data from different parts of the season 

 to show the daily and seasonal variations in the duration of the 

 trips which actually occurred in this experiment. 



A simple graphic representation of the average duration of the 

 trips and their variations in the course of a day is obtained by 

 plotting the progressive totals of all the bees which have left the 

 hive and of all those that have returned up to the time of each suc- 

 cessive reading from the cyclometers. The plotting of the exits 

 gives a continually ascending curve to the end of the day's flight. 

 The plotting of the returns gives a curve which follows the first 

 curve in time, as determined by the average duration of each trip 

 for that period of the day's flight. The horizontal distance between 

 the two curves represents the average time the bees spend in the field. 

 The vertical distance between the two curves at any time of the day 

 represents the number of bees in the field at that time. 



If we suppose a purely hypothetical case, in which a constant 

 number of bees leave the hive and return within a certain definite 

 interval of time throughout the day, two parallel Straight lines, 

 representing the exits and returns, would be obtained when plotted 

 as described above. By comparing two similar triangles on such a 

 diagram it is found that the ratio of the number of bees (n) which 

 return in a given interval of time (t), is to this time (t) as the 

 number of bees in the field (/) is to the average duration of the 



trip (x) ; that is, n:t::f:x y or a>= — • Turning now, for example, 



to Table 3 (May 15, fig. 4) and Table 4 (July 10, fig. 9) it is found 

 that the number of bees in the field increases at the beginning of 

 the day and diminishes as the day closes. Between these two periods 

 of the day there is a period during which the total number of bees 

 in the field at any particular time is reasonably near a constant num- 



